Face Generation¶

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data¶

You'll be using two datasets in this project:

MNIST
CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)

Found mnist Data
Found celeba Data

alpha = 0.02

Explore the Data¶

MNIST¶

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')

<matplotlib.image.AxesImage at 0x7fe1844959e8>

CelebA¶

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))

<matplotlib.image.AxesImage at 0x7fe18406bcc0>

Preprocess the Data¶

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network¶

You'll build the components necessary to build a GANs by implementing the following functions below:

model_inputs
discriminator
generator
model_loss
model_opt
train

Check the Version of TensorFlow and Access to GPU¶

This will check to make sure you have the correct version of TensorFlow and access to a GPU

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))

TensorFlow Version: 1.0.1
Default GPU Device: /gpu:0

Input¶

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
Z input placeholder with rank 2 using z_dim.
Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    img = tf.placeholder(tf.float32, shape=[None, image_width, image_height, image_channels], name='input_z')
    inputs_z = tf.placeholder(tf.float32, shape=[None, z_dim], name='input_z')
    lr = tf.placeholder(tf.float32, name='lr')
    return img, inputs_z, lr


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)

Tests Passed

Discriminator¶

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variabes in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the generator, tensor logits of the generator).

def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param image: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    
#     print('discriminator')
#     print('images: ', images)
    # TODO: Implement Function
    with tf.variable_scope('discriminator', reuse=reuse):
        # Input layer is ?x28x28x3
        x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same', 
                              kernel_initializer = tf.random_normal_initializer(stddev=0.02))
        relu1 = tf.maximum(0.02 * x1, x1)
        # 14x14x64
#         print(x1)
        x2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same', 
                              kernel_initializer = tf.random_normal_initializer(stddev=0.02))
        bn2 = tf.layers.batch_normalization(x2, training=True)
        relu2 = tf.maximum(0.02 * bn2, bn2)
        # 7x7x128
#         print(x2)
        x3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same', 
                              kernel_initializer = tf.random_normal_initializer(stddev=0.02))
        bn3 = tf.layers.batch_normalization(x3, training=True)
        relu3 = tf.maximum(0.02 * bn3, bn3)
        # 4x4x256 
#         print(x3)
        # Flatten it
        flat = tf.reshape(relu3, (-1, 4*4*256))
        logits = tf.layers.dense(flat, 1)
#         print(logits)
        out = tf.sigmoid(logits)
#         print('discriminator out: ', out)
        return out, logits



"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)

Tests Passed

Generator¶

Implement generator to generate an image using z. This function should be able to reuse the variabes in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
#     print('generator')
#     print('out_channel_dim: ', out_channel_dim)
    # TODO: Implement Function
    with tf.variable_scope('generator', reuse=not is_train):
        # First fully connected layer
        x1 = tf.layers.dense(z, 7*7*512)
        # Reshape it to start the convolutional stack
        x1 = tf.reshape(x1, (-1, 7, 7, 512))
#         x1 = tf.layers.batch_normalization(x1, training=training)
        x1 = tf.maximum(alpha * x1, x1)
        # 7x7x512 now
#         print(x1)
        x2 = tf.layers.conv2d_transpose(x1, 256, 5, strides=1, padding='same')
        x2 = tf.layers.batch_normalization(x2, training=is_train)
        x2 = tf.maximum(alpha * x2, x2)
        # 7x7x256 now
#         print(x2)
        x3 = tf.layers.conv2d_transpose(x2, 128, 5, strides=2, padding='same')
        x3 = tf.layers.batch_normalization(x3, training=is_train)
        x3 = tf.maximum(alpha * x3, x3)
        # 14x14x128 now
#         print(x3)

        x4 = tf.layers.conv2d_transpose(x3, 64, 5, strides=2, padding='same')
        x4 = tf.layers.batch_normalization(x4, training=is_train)
        x4 = tf.maximum(alpha * x4, x4)
        
        
        # Output layer
        logits = tf.layers.conv2d_transpose(x4, out_channel_dim, 5, strides=1, padding='same')
        # 28x28x3 now
#         print(logits)3
        out = tf.tanh(logits)
        
        return out


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)

Tests Passed

Loss¶

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

discriminator(images, reuse=False)
generator(z, out_channel_dim, is_train=True)

def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # TODO: Implement Function
    
#     print('Generator for fake images...')
    g_model = generator(input_z, out_channel_dim)
#     print('Passing discriminator with real images...')
    d_model_real, d_logits_real = discriminator(input_real)
#     print('Passing discriminator with fake images...')
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)

    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))

    d_loss = d_loss_real + d_loss_fake

    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)

Tests Passed

Optimization¶

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    
    # Get weights and bias to update
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    # Optimize
    d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
    
    update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
    g_updates = [opt for opt in update_ops if opt.name.startswith('generator')]
    with tf.control_dependencies(g_updates):
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)

Tests Passed

Neural Network Training¶

Show Output¶

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train¶

Implement train to build and train the GANs. Use the following functions you implemented:

model_inputs(image_width, image_height, image_channels, z_dim)
model_loss(input_real, input_z, out_channel_dim)
model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

# tf.reset_default_graph()
show_every=20
print_every=1

def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # TODO: Build Model
    
    print(epoch_count)
    print(batch_size)
    print(z_dim)
    print(learning_rate)
    print(beta1)
    print(data_shape)
    print(data_image_mode)
    
    

    image_width = data_shape[1]
    image_height = data_shape[2]
    image_channels = data_shape[3]
    input_real, input_z, learning_rate_ = model_inputs(image_width, image_height, image_channels, z_dim)
    d_loss, g_loss = model_loss(input_real, input_z, image_channels)
    d_train_opt, g_train_opt = model_opt(d_loss, g_loss, learning_rate, beta1)
    
    steps = 0
    print('Starting Session...')
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            print('Epoch :', epoch_i)
            for batch_images in get_batches(batch_size):
                # TODO: Train Model
                steps +=1
                
                batch_images = batch_images*2.0
                
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))

                _ = sess.run(d_train_opt, feed_dict={input_real: batch_images, input_z: batch_z})
                _ = sess.run(g_train_opt, feed_dict={input_z: batch_z})

                
                if steps % print_every == 0:
                    # At the end of each epoch, get the losses and print them out
                    train_loss_d = d_loss.eval({input_z: batch_z, input_real: batch_images})
                    train_loss_g = g_loss.eval({input_z: batch_z})

                    print("Epoch {}/{}...".format(epoch_i+1, epoch_count),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                    
                    
                    
                if steps % show_every == 0:
                    show_generator_output(sess, 16, input_z, image_channels, data_image_mode)

MNIST¶

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

batch_size = 256
z_dim = 30
learning_rate = 0.001
beta1 = 0.5
tf.reset_default_graph()
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)

2
256
30
0.001
0.5
(60000, 28, 28, 1)
L
Starting Session...
Epoch : 0
Epoch 1/2... Discriminator Loss: 3.3020... Generator Loss: 0.0466
Epoch 1/2... Discriminator Loss: 4.6159... Generator Loss: 9.0740
Epoch 1/2... Discriminator Loss: 6.1177... Generator Loss: 0.0035
Epoch 1/2... Discriminator Loss: 0.1113... Generator Loss: 9.5252
Epoch 1/2... Discriminator Loss: 0.3481... Generator Loss: 4.2335
Epoch 1/2... Discriminator Loss: 3.5535... Generator Loss: 0.0510
Epoch 1/2... Discriminator Loss: 3.8407... Generator Loss: 13.7251
Epoch 1/2... Discriminator Loss: 0.1166... Generator Loss: 4.5529
Epoch 1/2... Discriminator Loss: 2.9087... Generator Loss: 0.2041
Epoch 1/2... Discriminator Loss: 0.4441... Generator Loss: 12.9108
Epoch 1/2... Discriminator Loss: 0.6046... Generator Loss: 13.9347
Epoch 1/2... Discriminator Loss: 0.1395... Generator Loss: 11.7490
Epoch 1/2... Discriminator Loss: 0.0356... Generator Loss: 7.7641
Epoch 1/2... Discriminator Loss: 0.2007... Generator Loss: 2.4320
Epoch 1/2... Discriminator Loss: 0.8973... Generator Loss: 1.9409
Epoch 1/2... Discriminator Loss: 0.1649... Generator Loss: 7.3149
Epoch 1/2... Discriminator Loss: 0.3680... Generator Loss: 3.9573
Epoch 1/2... Discriminator Loss: 2.0208... Generator Loss: 1.2309
Epoch 1/2... Discriminator Loss: 0.9862... Generator Loss: 4.2856
Epoch 1/2... Discriminator Loss: 1.0985... Generator Loss: 1.9580

Epoch 1/2... Discriminator Loss: 0.7832... Generator Loss: 3.5349
Epoch 1/2... Discriminator Loss: 0.4457... Generator Loss: 1.6426
Epoch 1/2... Discriminator Loss: 0.2319... Generator Loss: 2.4024
Epoch 1/2... Discriminator Loss: 0.2639... Generator Loss: 2.6415
Epoch 1/2... Discriminator Loss: 0.2762... Generator Loss: 2.4362
Epoch 1/2... Discriminator Loss: 0.2490... Generator Loss: 3.3696
Epoch 1/2... Discriminator Loss: 0.1862... Generator Loss: 2.9051
Epoch 1/2... Discriminator Loss: 0.1432... Generator Loss: 3.0642
Epoch 1/2... Discriminator Loss: 0.1160... Generator Loss: 3.5178
Epoch 1/2... Discriminator Loss: 0.2090... Generator Loss: 2.4886
Epoch 1/2... Discriminator Loss: 0.4469... Generator Loss: 1.8499
Epoch 1/2... Discriminator Loss: 0.4798... Generator Loss: 6.7726
Epoch 1/2... Discriminator Loss: 0.5922... Generator Loss: 2.3582
Epoch 1/2... Discriminator Loss: 0.0987... Generator Loss: 5.6949
Epoch 1/2... Discriminator Loss: 0.1407... Generator Loss: 4.0548
Epoch 1/2... Discriminator Loss: 0.3169... Generator Loss: 1.9841
Epoch 1/2... Discriminator Loss: 0.2926... Generator Loss: 7.8862
Epoch 1/2... Discriminator Loss: 0.0831... Generator Loss: 3.8180
Epoch 1/2... Discriminator Loss: 1.1319... Generator Loss: 0.6162
Epoch 1/2... Discriminator Loss: 5.0390... Generator Loss: 14.5770

Epoch 1/2... Discriminator Loss: 0.0222... Generator Loss: 5.5399
Epoch 1/2... Discriminator Loss: 2.6529... Generator Loss: 0.2101
Epoch 1/2... Discriminator Loss: 0.1883... Generator Loss: 6.8056
Epoch 1/2... Discriminator Loss: 0.8175... Generator Loss: 7.1917
Epoch 1/2... Discriminator Loss: 0.4154... Generator Loss: 3.5105
Epoch 1/2... Discriminator Loss: 1.0952... Generator Loss: 1.2026
Epoch 1/2... Discriminator Loss: 0.6128... Generator Loss: 3.8570
Epoch 1/2... Discriminator Loss: 0.6874... Generator Loss: 2.0786
Epoch 1/2... Discriminator Loss: 1.1754... Generator Loss: 1.0518
Epoch 1/2... Discriminator Loss: 0.8810... Generator Loss: 4.8667
Epoch 1/2... Discriminator Loss: 0.3853... Generator Loss: 2.0764
Epoch 1/2... Discriminator Loss: 0.1382... Generator Loss: 2.8152
Epoch 1/2... Discriminator Loss: 0.1064... Generator Loss: 3.7178
Epoch 1/2... Discriminator Loss: 0.1045... Generator Loss: 3.7640
Epoch 1/2... Discriminator Loss: 0.0853... Generator Loss: 3.8893
Epoch 1/2... Discriminator Loss: 0.0864... Generator Loss: 3.8658
Epoch 1/2... Discriminator Loss: 0.0849... Generator Loss: 3.7184
Epoch 1/2... Discriminator Loss: 0.1103... Generator Loss: 3.1126
Epoch 1/2... Discriminator Loss: 0.0629... Generator Loss: 4.5111
Epoch 1/2... Discriminator Loss: 0.0541... Generator Loss: 4.1935

Epoch 1/2... Discriminator Loss: 0.0700... Generator Loss: 3.5581
Epoch 1/2... Discriminator Loss: 0.0488... Generator Loss: 4.7828
Epoch 1/2... Discriminator Loss: 0.0517... Generator Loss: 4.8373
Epoch 1/2... Discriminator Loss: 0.0379... Generator Loss: 4.6591
Epoch 1/2... Discriminator Loss: 0.0280... Generator Loss: 5.0325
Epoch 1/2... Discriminator Loss: 0.0206... Generator Loss: 5.4950
Epoch 1/2... Discriminator Loss: 0.0116... Generator Loss: 5.7959
Epoch 1/2... Discriminator Loss: 0.0131... Generator Loss: 5.9394
Epoch 1/2... Discriminator Loss: 0.0127... Generator Loss: 5.9572
Epoch 1/2... Discriminator Loss: 0.0115... Generator Loss: 5.6238
Epoch 1/2... Discriminator Loss: 0.0797... Generator Loss: 2.6877
Epoch 1/2... Discriminator Loss: 0.0200... Generator Loss: 7.6860
Epoch 1/2... Discriminator Loss: 0.0205... Generator Loss: 7.8717
Epoch 1/2... Discriminator Loss: 0.0473... Generator Loss: 4.0619
Epoch 1/2... Discriminator Loss: 0.6509... Generator Loss: 1.1557
Epoch 1/2... Discriminator Loss: 5.4671... Generator Loss: 19.3343
Epoch 1/2... Discriminator Loss: 7.4333... Generator Loss: 0.0160
Epoch 1/2... Discriminator Loss: 0.6470... Generator Loss: 4.0470
Epoch 1/2... Discriminator Loss: 1.4657... Generator Loss: 4.4853
Epoch 1/2... Discriminator Loss: 0.9598... Generator Loss: 1.6393

Epoch 1/2... Discriminator Loss: 0.3740... Generator Loss: 2.0341
Epoch 1/2... Discriminator Loss: 0.1859... Generator Loss: 3.2152
Epoch 1/2... Discriminator Loss: 0.1592... Generator Loss: 3.1240
Epoch 1/2... Discriminator Loss: 0.1792... Generator Loss: 2.6671
Epoch 1/2... Discriminator Loss: 0.2097... Generator Loss: 2.4291
Epoch 1/2... Discriminator Loss: 0.1831... Generator Loss: 2.5558
Epoch 1/2... Discriminator Loss: 0.2028... Generator Loss: 2.5367
Epoch 1/2... Discriminator Loss: 0.1667... Generator Loss: 2.6072
Epoch 1/2... Discriminator Loss: 0.2372... Generator Loss: 2.1171
Epoch 1/2... Discriminator Loss: 0.3401... Generator Loss: 1.6659
Epoch 1/2... Discriminator Loss: 0.6946... Generator Loss: 1.3285
Epoch 1/2... Discriminator Loss: 0.8671... Generator Loss: 1.4705
Epoch 1/2... Discriminator Loss: 0.6376... Generator Loss: 2.2369
Epoch 1/2... Discriminator Loss: 0.6756... Generator Loss: 1.5433
Epoch 1/2... Discriminator Loss: 0.5527... Generator Loss: 2.4222
Epoch 1/2... Discriminator Loss: 0.7617... Generator Loss: 1.1338
Epoch 1/2... Discriminator Loss: 0.9692... Generator Loss: 4.1960
Epoch 1/2... Discriminator Loss: 1.6885... Generator Loss: 0.3708
Epoch 1/2... Discriminator Loss: 2.0141... Generator Loss: 6.5250
Epoch 1/2... Discriminator Loss: 0.6618... Generator Loss: 1.5206

Epoch 1/2... Discriminator Loss: 0.6819... Generator Loss: 1.6444
Epoch 1/2... Discriminator Loss: 0.6461... Generator Loss: 2.7425
Epoch 1/2... Discriminator Loss: 0.8463... Generator Loss: 1.5887
Epoch 1/2... Discriminator Loss: 0.9968... Generator Loss: 1.3143
Epoch 1/2... Discriminator Loss: 1.1009... Generator Loss: 1.6059
Epoch 1/2... Discriminator Loss: 1.1747... Generator Loss: 1.3925
Epoch 1/2... Discriminator Loss: 1.1571... Generator Loss: 1.2570
Epoch 1/2... Discriminator Loss: 1.0960... Generator Loss: 1.6755
Epoch 1/2... Discriminator Loss: 0.8572... Generator Loss: 1.5593
Epoch 1/2... Discriminator Loss: 0.5750... Generator Loss: 2.0777
Epoch 1/2... Discriminator Loss: 0.5550... Generator Loss: 1.8461
Epoch 1/2... Discriminator Loss: 0.5812... Generator Loss: 1.5043
Epoch 1/2... Discriminator Loss: 0.8643... Generator Loss: 2.2207
Epoch 1/2... Discriminator Loss: 1.2791... Generator Loss: 0.5823
Epoch 1/2... Discriminator Loss: 2.7605... Generator Loss: 5.3152
Epoch 1/2... Discriminator Loss: 1.5183... Generator Loss: 0.3618
Epoch 1/2... Discriminator Loss: 0.6902... Generator Loss: 2.7520
Epoch 1/2... Discriminator Loss: 0.7282... Generator Loss: 1.5538
Epoch 1/2... Discriminator Loss: 1.3559... Generator Loss: 0.6432
Epoch 1/2... Discriminator Loss: 1.2317... Generator Loss: 3.0510

Epoch 1/2... Discriminator Loss: 0.8286... Generator Loss: 1.1394
Epoch 1/2... Discriminator Loss: 0.8544... Generator Loss: 1.1391
Epoch 1/2... Discriminator Loss: 1.0399... Generator Loss: 1.6826
Epoch 1/2... Discriminator Loss: 0.9845... Generator Loss: 1.2698
Epoch 1/2... Discriminator Loss: 0.9373... Generator Loss: 1.4766
Epoch 1/2... Discriminator Loss: 0.9219... Generator Loss: 0.9170
Epoch 1/2... Discriminator Loss: 0.9075... Generator Loss: 2.6835
Epoch 1/2... Discriminator Loss: 1.2234... Generator Loss: 0.5588
Epoch 1/2... Discriminator Loss: 1.1427... Generator Loss: 3.5889
Epoch 1/2... Discriminator Loss: 1.0881... Generator Loss: 0.7103
Epoch 1/2... Discriminator Loss: 0.7977... Generator Loss: 2.0060
Epoch 1/2... Discriminator Loss: 1.0110... Generator Loss: 1.0252
Epoch 1/2... Discriminator Loss: 0.8529... Generator Loss: 2.1366
Epoch 1/2... Discriminator Loss: 1.0959... Generator Loss: 0.7952
Epoch 1/2... Discriminator Loss: 1.4146... Generator Loss: 3.1724
Epoch 1/2... Discriminator Loss: 1.5835... Generator Loss: 0.4573
Epoch 1/2... Discriminator Loss: 1.1667... Generator Loss: 4.1910
Epoch 1/2... Discriminator Loss: 0.5266... Generator Loss: 1.7636
Epoch 1/2... Discriminator Loss: 0.6422... Generator Loss: 1.1595
Epoch 1/2... Discriminator Loss: 0.5781... Generator Loss: 2.4615

Epoch 1/2... Discriminator Loss: 0.8539... Generator Loss: 1.0697
Epoch 1/2... Discriminator Loss: 1.1319... Generator Loss: 1.0234
Epoch 1/2... Discriminator Loss: 1.1537... Generator Loss: 1.5216
Epoch 1/2... Discriminator Loss: 1.0482... Generator Loss: 0.6428
Epoch 1/2... Discriminator Loss: 0.9129... Generator Loss: 3.6048
Epoch 1/2... Discriminator Loss: 0.9575... Generator Loss: 0.9712
Epoch 1/2... Discriminator Loss: 0.9210... Generator Loss: 1.6598
Epoch 1/2... Discriminator Loss: 0.9383... Generator Loss: 1.4681
Epoch 1/2... Discriminator Loss: 1.0197... Generator Loss: 0.9833
Epoch 1/2... Discriminator Loss: 0.9128... Generator Loss: 1.9313
Epoch 1/2... Discriminator Loss: 1.2561... Generator Loss: 0.5910
Epoch 1/2... Discriminator Loss: 1.3890... Generator Loss: 4.3496
Epoch 1/2... Discriminator Loss: 1.0570... Generator Loss: 0.7496
Epoch 1/2... Discriminator Loss: 0.8727... Generator Loss: 1.6743
Epoch 1/2... Discriminator Loss: 1.0717... Generator Loss: 1.1562
Epoch 1/2... Discriminator Loss: 1.0320... Generator Loss: 1.5642
Epoch 1/2... Discriminator Loss: 1.3609... Generator Loss: 0.9412
Epoch 1/2... Discriminator Loss: 1.4085... Generator Loss: 2.2864
Epoch 1/2... Discriminator Loss: 2.1594... Generator Loss: 0.2185
Epoch 1/2... Discriminator Loss: 2.3724... Generator Loss: 4.6466

Epoch 1/2... Discriminator Loss: 1.2787... Generator Loss: 0.6336
Epoch 1/2... Discriminator Loss: 1.1009... Generator Loss: 1.0164
Epoch 1/2... Discriminator Loss: 1.2341... Generator Loss: 1.8459
Epoch 1/2... Discriminator Loss: 1.1372... Generator Loss: 0.7008
Epoch 1/2... Discriminator Loss: 0.9894... Generator Loss: 1.8009
Epoch 1/2... Discriminator Loss: 0.9391... Generator Loss: 0.9430
Epoch 1/2... Discriminator Loss: 0.9081... Generator Loss: 1.1708
Epoch 1/2... Discriminator Loss: 0.9687... Generator Loss: 1.5222
Epoch 1/2... Discriminator Loss: 1.1441... Generator Loss: 0.6132
Epoch 1/2... Discriminator Loss: 1.2154... Generator Loss: 2.8225
Epoch 1/2... Discriminator Loss: 1.4318... Generator Loss: 0.4081
Epoch 1/2... Discriminator Loss: 1.1954... Generator Loss: 2.1115
Epoch 1/2... Discriminator Loss: 1.2657... Generator Loss: 0.6579
Epoch 1/2... Discriminator Loss: 1.0929... Generator Loss: 1.7561
Epoch 1/2... Discriminator Loss: 0.8784... Generator Loss: 1.0637
Epoch 1/2... Discriminator Loss: 0.6652... Generator Loss: 1.9319
Epoch 1/2... Discriminator Loss: 0.8054... Generator Loss: 1.1885
Epoch 1/2... Discriminator Loss: 0.9508... Generator Loss: 1.4508
Epoch 1/2... Discriminator Loss: 1.1604... Generator Loss: 1.1740
Epoch 1/2... Discriminator Loss: 1.0114... Generator Loss: 1.3055

Epoch 1/2... Discriminator Loss: 0.9631... Generator Loss: 1.2951
Epoch 1/2... Discriminator Loss: 1.0283... Generator Loss: 1.4547
Epoch 1/2... Discriminator Loss: 1.0721... Generator Loss: 0.9890
Epoch 1/2... Discriminator Loss: 1.0205... Generator Loss: 1.4051
Epoch 1/2... Discriminator Loss: 0.7498... Generator Loss: 1.3785
Epoch 1/2... Discriminator Loss: 0.6861... Generator Loss: 1.4696
Epoch 1/2... Discriminator Loss: 0.5753... Generator Loss: 1.6829
Epoch 1/2... Discriminator Loss: 0.6717... Generator Loss: 1.5107
Epoch 1/2... Discriminator Loss: 0.8217... Generator Loss: 1.5281
Epoch 1/2... Discriminator Loss: 1.0225... Generator Loss: 0.8060
Epoch 1/2... Discriminator Loss: 2.6818... Generator Loss: 4.1134
Epoch 1/2... Discriminator Loss: 3.8503... Generator Loss: 0.0399
Epoch 1/2... Discriminator Loss: 1.1000... Generator Loss: 1.7804
Epoch 1/2... Discriminator Loss: 1.1282... Generator Loss: 1.9837
Epoch 1/2... Discriminator Loss: 1.3506... Generator Loss: 0.4708
Epoch 1/2... Discriminator Loss: 0.9388... Generator Loss: 1.6431
Epoch 1/2... Discriminator Loss: 1.0496... Generator Loss: 1.1062
Epoch 1/2... Discriminator Loss: 1.1414... Generator Loss: 0.8543
Epoch 1/2... Discriminator Loss: 1.2041... Generator Loss: 1.3360
Epoch 1/2... Discriminator Loss: 1.1236... Generator Loss: 0.8242

CelebA¶

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

batch_size = 128
z_dim = 100
learning_rate = 0.001
beta1 = 0.5
tf.reset_default_graph()


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)

1
128
100
0.001
0.5
(202599, 28, 28, 3)
RGB
Starting Session...
Epoch : 0
Epoch 1/1... Discriminator Loss: 14.5924... Generator Loss: 0.0000
Epoch 1/1... Discriminator Loss: 6.3558... Generator Loss: 0.0020
Epoch 1/1... Discriminator Loss: 0.7257... Generator Loss: 5.3335
Epoch 1/1... Discriminator Loss: 3.1278... Generator Loss: 0.0561
Epoch 1/1... Discriminator Loss: 1.2184... Generator Loss: 11.5896
Epoch 1/1... Discriminator Loss: 0.1711... Generator Loss: 4.7138
Epoch 1/1... Discriminator Loss: 4.9564... Generator Loss: 0.0085
Epoch 1/1... Discriminator Loss: 0.4976... Generator Loss: 13.2582
Epoch 1/1... Discriminator Loss: 0.5123... Generator Loss: 11.2189
Epoch 1/1... Discriminator Loss: 0.8531... Generator Loss: 0.9688
Epoch 1/1... Discriminator Loss: 0.2868... Generator Loss: 8.7854
Epoch 1/1... Discriminator Loss: 0.4293... Generator Loss: 2.8205
Epoch 1/1... Discriminator Loss: 0.7588... Generator Loss: 1.2598
Epoch 1/1... Discriminator Loss: 0.5866... Generator Loss: 8.9424
Epoch 1/1... Discriminator Loss: 0.2056... Generator Loss: 3.0496
Epoch 1/1... Discriminator Loss: 2.3297... Generator Loss: 0.7756
Epoch 1/1... Discriminator Loss: 0.9332... Generator Loss: 11.2198
Epoch 1/1... Discriminator Loss: 0.3686... Generator Loss: 10.5610
Epoch 1/1... Discriminator Loss: 0.0908... Generator Loss: 6.6074
Epoch 1/1... Discriminator Loss: 0.8008... Generator Loss: 1.6615

Epoch 1/1... Discriminator Loss: 0.1109... Generator Loss: 6.0966
Epoch 1/1... Discriminator Loss: 0.4407... Generator Loss: 2.8787
Epoch 1/1... Discriminator Loss: 0.5048... Generator Loss: 2.0791
Epoch 1/1... Discriminator Loss: 0.9217... Generator Loss: 1.9951
Epoch 1/1... Discriminator Loss: 0.8991... Generator Loss: 6.8787
Epoch 1/1... Discriminator Loss: 4.8124... Generator Loss: 0.0538
Epoch 1/1... Discriminator Loss: 2.9389... Generator Loss: 7.2482
Epoch 1/1... Discriminator Loss: 0.5339... Generator Loss: 3.0318
Epoch 1/1... Discriminator Loss: 2.4628... Generator Loss: 0.3958
Epoch 1/1... Discriminator Loss: 0.7320... Generator Loss: 2.7026
Epoch 1/1... Discriminator Loss: 0.8570... Generator Loss: 2.4318
Epoch 1/1... Discriminator Loss: 1.1954... Generator Loss: 0.9736
Epoch 1/1... Discriminator Loss: 0.8292... Generator Loss: 2.1379
Epoch 1/1... Discriminator Loss: 0.5550... Generator Loss: 2.2977
Epoch 1/1... Discriminator Loss: 0.9393... Generator Loss: 0.9327
Epoch 1/1... Discriminator Loss: 1.1774... Generator Loss: 2.2708
Epoch 1/1... Discriminator Loss: 2.8469... Generator Loss: 0.1462
Epoch 1/1... Discriminator Loss: 3.2004... Generator Loss: 7.3220
Epoch 1/1... Discriminator Loss: 0.7572... Generator Loss: 2.3470
Epoch 1/1... Discriminator Loss: 1.3777... Generator Loss: 0.5241

Epoch 1/1... Discriminator Loss: 0.2821... Generator Loss: 2.3551
Epoch 1/1... Discriminator Loss: 0.2502... Generator Loss: 4.0144
Epoch 1/1... Discriminator Loss: 0.1828... Generator Loss: 3.7729
Epoch 1/1... Discriminator Loss: 0.2054... Generator Loss: 2.3048
Epoch 1/1... Discriminator Loss: 0.3499... Generator Loss: 1.5098
Epoch 1/1... Discriminator Loss: 0.2043... Generator Loss: 2.8289
Epoch 1/1... Discriminator Loss: 0.5018... Generator Loss: 1.4831
Epoch 1/1... Discriminator Loss: 0.5987... Generator Loss: 1.9864
Epoch 1/1... Discriminator Loss: 1.0595... Generator Loss: 0.8472
Epoch 1/1... Discriminator Loss: 0.8857... Generator Loss: 4.2403
Epoch 1/1... Discriminator Loss: 1.6250... Generator Loss: 0.4860
Epoch 1/1... Discriminator Loss: 0.8364... Generator Loss: 4.5183
Epoch 1/1... Discriminator Loss: 1.5472... Generator Loss: 0.7651
Epoch 1/1... Discriminator Loss: 1.1992... Generator Loss: 2.8474
Epoch 1/1... Discriminator Loss: 0.7536... Generator Loss: 1.1440
Epoch 1/1... Discriminator Loss: 0.6027... Generator Loss: 1.5879
Epoch 1/1... Discriminator Loss: 0.6483... Generator Loss: 1.4237
Epoch 1/1... Discriminator Loss: 0.7926... Generator Loss: 1.1867
Epoch 1/1... Discriminator Loss: 1.0339... Generator Loss: 1.4126
Epoch 1/1... Discriminator Loss: 1.3764... Generator Loss: 0.5163

Epoch 1/1... Discriminator Loss: 2.6342... Generator Loss: 5.3194
Epoch 1/1... Discriminator Loss: 1.1550... Generator Loss: 0.5730
Epoch 1/1... Discriminator Loss: 0.7945... Generator Loss: 2.1870
Epoch 1/1... Discriminator Loss: 0.7275... Generator Loss: 1.3842
Epoch 1/1... Discriminator Loss: 0.9055... Generator Loss: 0.9017
Epoch 1/1... Discriminator Loss: 0.6303... Generator Loss: 1.5828
Epoch 1/1... Discriminator Loss: 0.7088... Generator Loss: 1.2276
Epoch 1/1... Discriminator Loss: 0.6029... Generator Loss: 1.5961
Epoch 1/1... Discriminator Loss: 0.9606... Generator Loss: 0.8269
Epoch 1/1... Discriminator Loss: 1.2335... Generator Loss: 2.3773
Epoch 1/1... Discriminator Loss: 2.5848... Generator Loss: 0.1742
Epoch 1/1... Discriminator Loss: 2.0545... Generator Loss: 3.8457
Epoch 1/1... Discriminator Loss: 0.8206... Generator Loss: 1.0615
Epoch 1/1... Discriminator Loss: 1.4100... Generator Loss: 0.3930
Epoch 1/1... Discriminator Loss: 1.3665... Generator Loss: 2.2416
Epoch 1/1... Discriminator Loss: 1.3466... Generator Loss: 0.7740
Epoch 1/1... Discriminator Loss: 0.9463... Generator Loss: 2.3546
Epoch 1/1... Discriminator Loss: 1.5267... Generator Loss: 0.3910
Epoch 1/1... Discriminator Loss: 1.2434... Generator Loss: 3.6597
Epoch 1/1... Discriminator Loss: 1.0860... Generator Loss: 1.1483

Epoch 1/1... Discriminator Loss: 1.1890... Generator Loss: 0.7982
Epoch 1/1... Discriminator Loss: 1.3139... Generator Loss: 4.1582
Epoch 1/1... Discriminator Loss: 0.9566... Generator Loss: 1.0015
Epoch 1/1... Discriminator Loss: 0.8312... Generator Loss: 1.0722
Epoch 1/1... Discriminator Loss: 1.0399... Generator Loss: 1.5296
Epoch 1/1... Discriminator Loss: 1.8191... Generator Loss: 0.3576
Epoch 1/1... Discriminator Loss: 1.9012... Generator Loss: 3.0111
Epoch 1/1... Discriminator Loss: 1.5277... Generator Loss: 0.3939
Epoch 1/1... Discriminator Loss: 0.8972... Generator Loss: 1.1230
Epoch 1/1... Discriminator Loss: 1.2666... Generator Loss: 1.1941
Epoch 1/1... Discriminator Loss: 1.0940... Generator Loss: 1.1714
Epoch 1/1... Discriminator Loss: 1.0318... Generator Loss: 1.1384
Epoch 1/1... Discriminator Loss: 1.4663... Generator Loss: 0.6958
Epoch 1/1... Discriminator Loss: 1.8280... Generator Loss: 1.6977
Epoch 1/1... Discriminator Loss: 1.8326... Generator Loss: 0.2765
Epoch 1/1... Discriminator Loss: 1.1627... Generator Loss: 2.8889
Epoch 1/1... Discriminator Loss: 0.7903... Generator Loss: 1.9882
Epoch 1/1... Discriminator Loss: 0.7775... Generator Loss: 1.1448
Epoch 1/1... Discriminator Loss: 0.7110... Generator Loss: 1.1804
Epoch 1/1... Discriminator Loss: 0.6394... Generator Loss: 1.4319

Epoch 1/1... Discriminator Loss: 0.9483... Generator Loss: 0.9197
Epoch 1/1... Discriminator Loss: 1.1844... Generator Loss: 0.7246
Epoch 1/1... Discriminator Loss: 1.3090... Generator Loss: 0.9195
Epoch 1/1... Discriminator Loss: 1.0901... Generator Loss: 2.0573
Epoch 1/1... Discriminator Loss: 1.3406... Generator Loss: 0.4858
Epoch 1/1... Discriminator Loss: 1.0377... Generator Loss: 1.3236
Epoch 1/1... Discriminator Loss: 1.4709... Generator Loss: 0.7003
Epoch 1/1... Discriminator Loss: 1.1467... Generator Loss: 1.9093
Epoch 1/1... Discriminator Loss: 1.5774... Generator Loss: 0.4334
Epoch 1/1... Discriminator Loss: 1.2696... Generator Loss: 1.6850
Epoch 1/1... Discriminator Loss: 1.7965... Generator Loss: 0.4090
Epoch 1/1... Discriminator Loss: 1.4482... Generator Loss: 0.6600
Epoch 1/1... Discriminator Loss: 1.2103... Generator Loss: 1.1540
Epoch 1/1... Discriminator Loss: 1.1892... Generator Loss: 0.9036
Epoch 1/1... Discriminator Loss: 1.2725... Generator Loss: 0.8104
Epoch 1/1... Discriminator Loss: 1.5621... Generator Loss: 0.6346
Epoch 1/1... Discriminator Loss: 1.3174... Generator Loss: 1.7701
Epoch 1/1... Discriminator Loss: 1.2949... Generator Loss: 0.7212
Epoch 1/1... Discriminator Loss: 1.3738... Generator Loss: 1.4041
Epoch 1/1... Discriminator Loss: 1.6061... Generator Loss: 0.4641

Epoch 1/1... Discriminator Loss: 1.5069... Generator Loss: 1.6832
Epoch 1/1... Discriminator Loss: 1.3409... Generator Loss: 0.7353
Epoch 1/1... Discriminator Loss: 1.0119... Generator Loss: 1.1942
Epoch 1/1... Discriminator Loss: 0.9279... Generator Loss: 0.8734
Epoch 1/1... Discriminator Loss: 0.8295... Generator Loss: 1.0976
Epoch 1/1... Discriminator Loss: 0.7823... Generator Loss: 1.1404
Epoch 1/1... Discriminator Loss: 0.9376... Generator Loss: 1.0869
Epoch 1/1... Discriminator Loss: 1.5200... Generator Loss: 0.3764
Epoch 1/1... Discriminator Loss: 1.6852... Generator Loss: 1.9186
Epoch 1/1... Discriminator Loss: 1.7905... Generator Loss: 0.3460
Epoch 1/1... Discriminator Loss: 1.3418... Generator Loss: 0.7456
Epoch 1/1... Discriminator Loss: 1.0480... Generator Loss: 1.6030
Epoch 1/1... Discriminator Loss: 0.9840... Generator Loss: 0.7049
Epoch 1/1... Discriminator Loss: 0.8691... Generator Loss: 0.8876
Epoch 1/1... Discriminator Loss: 0.7141... Generator Loss: 1.5319
Epoch 1/1... Discriminator Loss: 1.1977... Generator Loss: 0.6638
Epoch 1/1... Discriminator Loss: 1.3858... Generator Loss: 1.4098
Epoch 1/1... Discriminator Loss: 1.6378... Generator Loss: 0.5446
Epoch 1/1... Discriminator Loss: 2.0209... Generator Loss: 0.3871
Epoch 1/1... Discriminator Loss: 2.0488... Generator Loss: 2.0618

Epoch 1/1... Discriminator Loss: 1.5212... Generator Loss: 0.4717
Epoch 1/1... Discriminator Loss: 1.6155... Generator Loss: 0.4710
Epoch 1/1... Discriminator Loss: 1.5082... Generator Loss: 1.6013
Epoch 1/1... Discriminator Loss: 1.4487... Generator Loss: 0.6435
Epoch 1/1... Discriminator Loss: 1.1456... Generator Loss: 0.9548
Epoch 1/1... Discriminator Loss: 1.2613... Generator Loss: 0.9076
Epoch 1/1... Discriminator Loss: 1.3549... Generator Loss: 0.7355
Epoch 1/1... Discriminator Loss: 1.2374... Generator Loss: 0.9655
Epoch 1/1... Discriminator Loss: 1.2181... Generator Loss: 0.9286
Epoch 1/1... Discriminator Loss: 1.2599... Generator Loss: 0.6942
Epoch 1/1... Discriminator Loss: 1.2826... Generator Loss: 1.0177
Epoch 1/1... Discriminator Loss: 1.4682... Generator Loss: 0.5032
Epoch 1/1... Discriminator Loss: 1.4338... Generator Loss: 2.1657
Epoch 1/1... Discriminator Loss: 1.3259... Generator Loss: 0.5413
Epoch 1/1... Discriminator Loss: 1.0648... Generator Loss: 1.2673
Epoch 1/1... Discriminator Loss: 0.9351... Generator Loss: 1.0229
Epoch 1/1... Discriminator Loss: 0.7646... Generator Loss: 1.0643
Epoch 1/1... Discriminator Loss: 0.7584... Generator Loss: 1.5718
Epoch 1/1... Discriminator Loss: 0.8351... Generator Loss: 0.9138
Epoch 1/1... Discriminator Loss: 0.8301... Generator Loss: 1.1544

Epoch 1/1... Discriminator Loss: 0.6970... Generator Loss: 1.5988
Epoch 1/1... Discriminator Loss: 0.8231... Generator Loss: 0.9287
Epoch 1/1... Discriminator Loss: 0.6212... Generator Loss: 1.5755
Epoch 1/1... Discriminator Loss: 0.7264... Generator Loss: 1.0009
Epoch 1/1... Discriminator Loss: 0.7555... Generator Loss: 1.5340
Epoch 1/1... Discriminator Loss: 0.8760... Generator Loss: 0.9499
Epoch 1/1... Discriminator Loss: 1.2632... Generator Loss: 1.5398
Epoch 1/1... Discriminator Loss: 2.4664... Generator Loss: 0.1463
Epoch 1/1... Discriminator Loss: 5.9560... Generator Loss: 6.8315
Epoch 1/1... Discriminator Loss: 1.5269... Generator Loss: 2.0576
Epoch 1/1... Discriminator Loss: 3.1844... Generator Loss: 0.0665
Epoch 1/1... Discriminator Loss: 1.5028... Generator Loss: 0.4209
Epoch 1/1... Discriminator Loss: 1.7231... Generator Loss: 2.1578
Epoch 1/1... Discriminator Loss: 1.5133... Generator Loss: 1.1581
Epoch 1/1... Discriminator Loss: 1.7662... Generator Loss: 0.4614
Epoch 1/1... Discriminator Loss: 1.4402... Generator Loss: 1.3543
Epoch 1/1... Discriminator Loss: 1.1322... Generator Loss: 1.3599
Epoch 1/1... Discriminator Loss: 1.1819... Generator Loss: 0.8134
Epoch 1/1... Discriminator Loss: 1.2006... Generator Loss: 0.8956
Epoch 1/1... Discriminator Loss: 1.2983... Generator Loss: 0.7984

Epoch 1/1... Discriminator Loss: 1.1134... Generator Loss: 0.8304
Epoch 1/1... Discriminator Loss: 1.0337... Generator Loss: 0.7441
Epoch 1/1... Discriminator Loss: 0.8532... Generator Loss: 0.9769
Epoch 1/1... Discriminator Loss: 0.9276... Generator Loss: 1.0134
Epoch 1/1... Discriminator Loss: 1.1997... Generator Loss: 0.7468
Epoch 1/1... Discriminator Loss: 1.1363... Generator Loss: 1.1833
Epoch 1/1... Discriminator Loss: 1.2408... Generator Loss: 0.6618
Epoch 1/1... Discriminator Loss: 1.1388... Generator Loss: 1.1146
Epoch 1/1... Discriminator Loss: 1.2371... Generator Loss: 0.7060
Epoch 1/1... Discriminator Loss: 1.1303... Generator Loss: 1.1248
Epoch 1/1... Discriminator Loss: 1.0464... Generator Loss: 0.7650
Epoch 1/1... Discriminator Loss: 0.9664... Generator Loss: 1.0932
Epoch 1/1... Discriminator Loss: 1.2366... Generator Loss: 0.7264
Epoch 1/1... Discriminator Loss: 1.5438... Generator Loss: 0.5955
Epoch 1/1... Discriminator Loss: 1.7926... Generator Loss: 0.4747
Epoch 1/1... Discriminator Loss: 2.2827... Generator Loss: 2.4600
Epoch 1/1... Discriminator Loss: 2.0614... Generator Loss: 0.2352
Epoch 1/1... Discriminator Loss: 1.4595... Generator Loss: 2.0071
Epoch 1/1... Discriminator Loss: 1.3589... Generator Loss: 0.7928
Epoch 1/1... Discriminator Loss: 1.4898... Generator Loss: 0.5747

Submitting This Project¶

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.